A week of KDE4 usage

[Duncan]

Alex Schuster posted on Tue, 10 May 2011 01:49:49 +0200 as excerpted:

> [Oh, this has become rather lengthy. It's a description of my various
> problems with KDE4, the details are not so important, no need to read it
> all. My question is: Are your experiences similar to mine?]

I read it all, because I'm interested in such things, but I snipped the 
list, as replying to all the individual items wasn't the point.

> I'm somewhat diesappointed with KDE4. I'm using it since 4.2, and it's
> become much much better - but still, there are just so many bugs. Is it
> just me, or it this normal? Would you suggest other people (being
> unskilled uses, not hackers) to use KDE4? What OS and desktop
> environment does your Mom's PC run?

With kde4 it has been made /very/ apparent that the kde devs simply aren't 
interested in creating a "just works" desktop for ordinary users, or even 
ordinary /power/ users.  If they were, they would NOT have insisted 4.2 
was ready for ordinary users, when it /very/ clearly was still alpha 
quality at best -- huge gaps of missing functionality, and bugs with the 
devs saying that's not implemented yet, so it's not as if they weren't 
/aware/ of the problem, what was there often broken, good as a technology 
preview, but not for use by people actually wanting to get stuff done.  
Yet they INSISTED it was ready for ordinary users!  And were they 
interested in normal users, they'd have not dropped support for the stable 
kde3 at the same time, when kde4 clearly wasn't ready.

So it's quite clear that the kde folk would rather simply have people who 
just want things to work, move on to other desktop environments and quit 
bothering the kde folks.

The same message comes thru when one considers the official name change, 
from kde, to "The KDE Software Collection".  "Nobody"'s going to use that 
in normal usage, or even the shorter KDE-SC, for the same reason "nobody" 
uses GNU/Linux in normal usage -- it's too long and inconvenient to say, 
however correct it arguably may be.  But further, looking at the PR for 
the name change, it again becomes apparent that they're simply no longer 
targeting the ordinary end user, but rather, now, they consider 
developers, etc, their target.  I remember wondering about the name 
change, until after reading the announcement, it hit me -- the were no 
longer focused on the ordinary end user and the name change, adding 
"Software Collection" was aimed at making the platform more inviting to 
developers, etc, who could build upon that collection, using it as a basis 
for their own apps and possibly their own platforms, etc.

Now, arguably (and I've repeatedly made this point myself so I obviously 
argue it to be so), by later 4.5, say 4.5.4 and 4.5.5, KDE (um... KDE-SC) 
had improved to the point that it was ready for what /should/ have been 
the official 4.0 release -- had it been aimed at ordinary users.  But 
again, it's NOT aimed at ordinary users any more.  So what they called 
4.0, which they clearly labeled as developer-only, freezing the libraries 
but with a barely functional UI skeleton over top, really /WAS/ a .0 
release IF YOU'RE TARGETING THE DEVELOPER AUDIENCE FOR WHOM A LIBRARY 
FREEZE IS SIGNIFICANT.

Looked at thru that filter, the filter of who their actual target audience 
is, now, all the rest begins to fall into place and make a WHOLE lot more 
sense!

Now, it's possible some will argue with that viewpoint.  But, I'd simply 
point at the facts.  How /else/ can they be reasonably interpreted?  Given 
the facts on the ground, the actual actions, tt /certainly/ makes more 
sense for KDE (KDE-SC) to be targeting developers than it does for them to 
be targeting ordinary users.  There may be other explanations, but I've 
yet to see any others that even remotely matches the facts as good as this 
one does.

You and I are both Gentoo users.  I make it a point to read the Gentoo-dev 
list, and occasionally find myself pointing out yet again that Gentoo has 
never been about hand-holding/baby-sitting the user.  Gentoo does try to 
provide good documentation, but if there's a choice between making a 
choice available that a user could hurt themselves with if they don't heed 
the news pre-warnings and post-install log-warnings, and Gentoo main site 
announcements, and... and..., and hiding that choice to protect the user 
at the expense of making Gentoo far less flexible (tho there are certain 
limits), time and again, Gentoo chooses to make that choice available, and 
if the user breaks their system because they couldn't follow the clear 
warnings and instructions, well, they get to keep the pieces.

In that regard Gentoo's nothing like the hand-holding distros that Ubuntu 
and the like try to be, nor SHOULD we be, nor SHOULD WE TRY to be.  Gentoo 
has its own niche, and fills it quite well.  It's not for everyone and we 
don't pretend that it is.  If people find it too difficult, etc, well, 
there's other distros out there, and we're happy enough to point them to a 
few of them.

KDE has always been that way to some degree as well.  It has always 
exposed way more config options than the "protect the user from his stupid 
self" Gnome, for instance, and has regularly gotten called for "all those 
confusing options".

Fine, there are other desktops out there.  If a user just wants it to work 
and doesn't want to be confused or scared by all those options, that's 
fine.  There's desktops for that.  KDE has never been one of them.  FWIW, 
this is what all the folks who want a unified "One true desktop Linux 
experience" miss.  The GNOME folks are HAPPY not to have the KDE guys 
there adding all those confusing options, while the KDE guys (both devs 
and users, BTW) are HAPPY to refer people who want to take away the 
ability to so heavily customize, over to the GNOME folks.  If they were 
forced to make a single desktop, they'd always be fighting each other and 
little would get done. People would drop out as a result, and go do 
something else (like create another alternative desktop) with their after 
all volunteer time.  This is part of the POWER of FLOSS, *NOT* the 
weakness so many see it as.

In a way, this evolution of KDE could be seen as taking that philosophy to 
the next level.  KDE will continue on, pressing into new territory, 
perhaps never really stabilizing things, but that's fine.  They're a 
"software collection" targeted at the developer that knows how to deal 
with such things.

And... I'll probably be one of those who sticks with them, because while 
I'm not a dev, I'm computer oriented enough to like the customization kde 
offers, and to take the bugs, now that they've gotten kde4 to a generally 
usable state, in stride.

OTOH, I've not seen nearly the number of bugs you apparently have.  We may 
use the same distro, but my usage patterns and hardware are obviously 
different.  And the bugs I DO see, I've been able to work around, or, in 
the first time for kde4, I've actually git-bisect a bug down to a specific 
commit, file the bug both upstream with kde and with gentoo, and using 
that commit to generate a reversing patch, I now have that applied to the
kde-base/plasma-workspace-4.6.3 I have presently installed.  I expect 
that'll continue working thru the 4.6 series if they don't fix it, and by 
4.7, hopefully, it /will/ be fixed, or if not, hopefully the patch will 
continue to apply, possibly with a tweak or two.

Meanwhile, the gap left by the former KDE, kde3, the one with all those 
power-user options but at the same time having matured into something 
reasonably stable, was pretty large.  The Trinity project has taken up 
where KDE dropped it, and is now porting what was kde3 to qt4 (since qt3 
is long EOLed by its people).  Will they be successful enough at it to 
make it the big-two (kde/gnome) into the big-three (adding trinity)?  
Chances are they won't.  But there's a much better chance they'll turn the 
big-four (including xfce and lxde) into the big-five, or the big-five 
(including enlightenment) into the big-six.  The big question is at what 
point are they big enough for the distributions to start carrying them?   
Because if they fail to reach that, then they'll almost certainly remain a 
minor footnote at best.  The /problem/ is that being a full desktop not 
just a window manager with a few compatible toys to go with it, they're 
rather bigger package-size-wise than the other relative minors, the *box 
family, icewm, fvwm, and the like.  As such, that's a rather larger 
investment for the distros to make, to ship and support all the packages 
that make up trinity, the former kde3.  So the bar to having trinity 
shipped by the distros is rather higher.  (Additionally, there's likely to 
be some political fallout particularly in Islamic countries, etc, based on 
the name... unless they change it, of course, but to what?  Krusader, the 
dual-pane kde file manager, must have had similar issues, possibly keeping 
it out of contention for inclusion in kde proper, for political/religious 
reasons.)  But they NEED to get there, because that's how one gets the 
continual influx of new users discovering the software, keeping it viable 
longer term, after the interest of the original devs ultimately fades.

As to the what does your mom run question...

I'm a confirmed Linux-head, but I realized some time ago that I don't make 
a very good Linux missionary for the masses, because I'm far too tech (and 
freedom) oriented.  As you mentioned was your experience with kde4, not 
being able to recommend it to others based on what they see you dealing 
with, that's pretty much what I've found to be the case in general with 
Linux.  I *CHOOSE* to run Gentoo.  It takes me HOURS to build install a 
KDE update that binary distributions install in fractions of an hour.  I 
won't (legally can't, because I can't agree to the EULAs, which are 
binding enough here in the US to at minimum be a legal issue) install 
servantware.  As such, no flash (gnash works sometimes, sometimes not, 
youtube and certain other sites there's downloaders for, and I can play 
the videos in smplayer, vlc, or whatever), no nvidia or ati proprietary 
drivers, etc.  My dad sends me *.ppt files from time to time that I don't 
have the software installed to open.  Sometimes I spend weeks tracing down 
a kernel bug in a prerelease kernel I'm running... etc.

So while I'd /like/ to be a Linux missionary, in general, I don't bother, 
because most of the time I'd be rather convincing people that it's not for 
them.  I /could/ learn how Ubuntu works, etc, and thus be a rather more 
effective missionary, but that's not interesting to me, so I don't.

My mom doesn't do computers much, but my dad does reasonably well for a 
guy in his 70s that didn't work with computers in his career.  However, 
they live several states away and I'm not as close to them as I might be.

Fortunately, my sister and her family live next door, and she's quite good 
with computers, using them in her publishing job.  But AFAIK, it's all 
servantware, and that's what my dad uses, too.  That used to bother me, 
but the fact is, I'm /not/ a particularly good missionary for Linux, and 
being as far from them as I am, I finally decided it's best just to let 
them be with what they know.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman

___________________________________________________
This message is from the kde mailing list.
Account management:  https://mail.kde.org/mailman/listinfo/kde.
Archives: http://lists.kde.org/.
More info: http://www.kde.org/faq.html.